Project Summary In the US, one in seven adults has type 2 diabetes (T2D), and if current trends continue, it is projected that T2D prevalence can rise to as high as one in three adults by the year 2050. Furthermore, the burden of this health condition disproportionately affects US ethnic minorities; whereas the current T2D prevalence is 11% among Whites, it is 22% among Blacks and Hispanics. Obesity is a strong risk factor for T2D and part of this effect is hypothesized to be due to the development of non-alcoholic fatty liver disease (NAFLD). However, the extent to which adiposity acts on the development of T2D through liver fat is unknown. Current race/ethnic-specific guidelines for the identification of obesity using body anthropometry (e.g. body mass index) are established for White and Asian populations. However, race/ethnic-specific guidelines are lacking for US Black and Hispanic populations, primarily because evidence has been limited to findings from cross-sectional studies. We propose to test the hypothesis that the obesity-T2D relationship will be evident at different clinically relevant anthropometric values among Blacks and Hispanics compared to Whites. In addition, we propose to estimate the mediating effect of liver fat on the association between anthropometry and T2D. The proposed investigation leverages the rich longitudinal data from the Multi-Ethnic Study of Atherosclerosis (MESA), a cohort study of 6,814 White, Black, Hispanic and Chinese adults that have been followed since the year 2000. Baseline data will identify obesity using several anthropometric measures, and to identify the degree of fatty liver using Computed Tomography scans. Individuals were followed for incident T2D in four subsequent follow-up exam visits. Proportional hazards regression will be used to estimate race-specific associations between anthropometry and incident T2D; multivariable linear regression analyses will be used to evaluate race- specific associations between anthropometric measurements and liver fat at baseline; and we will use inverse probability weighted proportional hazards regression to estimate the mediating effect of obesity on T2D through liver fat. Prior studies of the association of anthropometric measures and T2D in Black and Hispanic populations were limited by their cross-sectional study design, or lacked multi-ethnic groups to evaluate heterogeneity of the T2D risk between race/ethnic groups. In contrast, our proposal is innovative by evaluating this association in a large, well-characterized prospective cohort of multi-racial/ethnic groups. The significance of the proposal is reflected in the much higher prevalence of T2D in the understudied Black and Hispanic populations. The interdisciplinary training environment, expert mentorship in the fields of obesity and T2D epidemiology, health disparities and epidemiologic methods, will provide the applicant a platform in which he can strengthen his training goals and contribute to the field of chronic disease as a future independent researcher in T2D prevention epidemiology. !